Agentic Software Engineer (f/m/d)

In this role you will design and build agentic tooling and a robust verification layer to govern autonomous code changes. You’ll work across cross-functional teams to integrate orchestration, context management, and evaluation harnesses for agentic developers, while also hands-on refining a Java/Spring codebase. You help advance Sonar’s platform as the backbone of AI-driven software development, contributing to architecture and evaluating emerging agent frameworks. This is a fast-paced, impact-driven opportunity in Bochum.

• Pension Scheme: 1st Pillar (Unterstützungskasse): automatic contribution by Sonar, 3% of gross salary
• Pension Scheme: 2nd Pillar (bAV): voluntary, 15% contribution by Sonar
• 28 PTO days for Geneva-region employees
• Public transport reimbursement 60%
• Generous discretionary Company Growth Bonus
• Global workforce across 20+ countries

• Design, build, and iterate on agentic tooling: orchestration, tool-use, context management, and evaluation harnesses for autonomous code-modification agents
• Develop and refine the verification layer to gate agent outputs: automated test generation, regression detection, static analysis (within SonarQube), and human-in-the-loop checks
• Spend ~30% of time reviewing agent-generated pull requests, fixing issues, and feeding lessons back into tooling
• Collaborate with Server, Cloud Platform, and value-stream squads during migrations to earn trust and alignment
• Contribute to Architecture Decision Records (ADRs) and standards for the Unified Core
• Monitor and assess emerging agent frameworks to identify practical, value-adding innovations

• Hands-on experience building or operating multi-agent or LLM-agent systems in production
• Experience building evaluation or verification systems for AI-generated changes (e.g., automated testing, static analysis, regression detection)
• Strong software engineering fundamentals in Java and Spring (or comparable JVM/enterprise stack)
• Proven ability to own complex technical problems independently in an R&D-flavored production environment
• Experience influencing engineers and squads you don’t manage

• Independent problem-solving
• Influencing and cross-functional collaboration
• Clear communication and stakeholder management
• Multi-agent/LLM-agent systems in production
• Evaluation/verification systems for AI-generated changes
• Java and Spring (or JVM/enterprise stack)